The vast majority of glioblastoma (GBM) patients will develop tumor recurrence, despite being treated with near-total resection and post-operative radiation/chemotherapy. The local infiltration of the brain is a major factor in the failure of current treatment modalities, and the mechanisms underlying infiltrating behavior of tumors cells and treatment resistance are still poorly understood. To date, most anti-cancer therapies aim to eliminate rapidly proliferating tumor cells. Thus, the novel discovery of slow-cycling, radioresistant CD133+ GBM stem-like cells possessing the enhanced ability to repopulate tumors provides a new model to explain our inability to eradicate tumors. Our data indicated that CD133+ GBM cells isolated from recurrent GBM tumors are actively migrating cells and can migrate out from tumor spheres that they initiated in culture;additionally, intracranial injecting CD133+ GBM cells can develop infiltrating tumors that display extensive hypervascularity and pseudopalisading necrosis. The transcription profiles of CD133+ GBM cells further explored a series of the molecular signatures of neural crest cells and radial glial cells, as well as a group of genes involved in regulating cell migration, suggesting inherently migratory properties of the recurrent GBM tumor of origin. In this application, we aim to use these functional features and molecular information of CD133+ GBM stem cells to identify genes and pathway required for such an aggressively migratory function, which is believed to be a prerequisite for giving rise to secondary tumor foci. We hypothesize that target internal cues that promote the GBM stem cell migrating will restrain the tumor cell infiltrating the adjacent brain tissue. We propose two specific aims to test this hypothesis: 1. to screen and identify genes that promote CD133+ GBM stem cell migration. This aim will be carried out and accomplished by performing a loss-of-function gene screens, using an RNA interference library screening approach. Genes of interest can then be identified by the detection of gene knockdown mediated suppression of GBM sphere migrating out and growing into an adherent monolayer in vitro. 2. To test whether stable knockdown of cell migrating genes in CD133+ GBM stem cells will influence the infiltrating nature of GBM tumor that they initiated in mouse brain. We will generate lentivirus (LV) express short hairpin (sh) RNA that target the selected cell migrating genes identified in specific AIM 1. If a defined LVshRNA that knockdowns the expression of a particular gene leads to the suppression of GBM cell migration out of GBM spheres, the impact of gene knockdown in CD133+ GBM cells on developing an infiltrating tumor in mouse brain will be evaluated. Our ultimate goal is to untangle the gene pathway(s) that confer the infiltrating nature of malignant GBM tumor, thereby developing a novel therapeutic strategy to block tumor cells from infiltrating the normal brain tissue. PUBLIC HEALTH RELEVANCE: Glioblastoma stem cells (GSC) isolated from recurrent tumors are actively migrating cells, which are capable of self-renewing and giving rise to secondary tumors that are not treatable by radiation or chemotherapy. Our research goal is to identify genes that regulate GSC migration via a small interfering RNA (siRNA) screening approach. Our ultimate goal is to develop a gene pathway-targeted treatment protocol that can specifically block GSC migration and prevent GSC-mediated tumor recurrence.